Human Resources (HR) Conversations Dataset

This dataset contains 100 realistic HR support conversations between employees and HR representatives. Each dialogue is 11 turns long and covers common workplace issues such as HR portal access problems, payslip generation, onboarding questions, and other HR-related inquiries. It is ideal for training AI assistants, chatbots, and support models for internal HR departments.

Dataset Structure

Each conversation is stored as a JSON object with the following fields:

• id: Unique identifier for the conversation.
• domain: Always "HR / onboarding" for this dataset.
• problem: Short description of the employee's issue.
• customer_type: Employee persona (e.g., "confused beginner", "impatient executive").
• dialogue: List of alternating employee/HR messages. Each message has:
  • role: "customer" (employee) or "agent" (HR representative).
  • text: The message content.
• resolution: One-sentence summary of how the issue was resolved.

Example Conversation

{
  "id": 1,
  "domain": "HR / onboarding",
  "problem": "can't access HR portal",
  "customer_type": "confused beginner",
  "dialogue": [
    {
      "role": "customer",
      "text": "Hi, I just started this week and I can't seem to log into the HR portal. I've tried my credentials from the welcome email but nothing works."
    },
    {
      "role": "agent",
      "text": "Welcome to the team! I'm sorry for the trouble. Let me sort this out. Can I have your employee ID?"
    },
    {
      "role": "customer",
      "text": "My employee ID is EMP-44821. I started on Monday."
    },
    {
      "role": "agent",
      "text": "Thank you. I've found your record. I can see your portal account was created but the activation email expired before you clicked it. I'll reset it now."
    },
    {
      "role": "customer",
      "text": "Oh, I did get an email but wasn't sure what it was for. I think I missed it."
    },
    {
      "role": "agent",
      "text": "No problem at all — this happens frequently with new starters. I'm sending a fresh activation link to your work email right now."
    },
    {
      "role": "customer",
      "text": "I received it. Should I click 'Activate Account' now?"
    },
    {
      "role": "agent",
      "text": "Yes, click that button and set your new password. Password must be at least 10 characters with one number and one special character."
    },
    {
      "role": "customer",
      "text": "Done! I'm logged in now. I can see my profile, payslips, and leave balance. Everything is there!"
    },
    {
      "role": "agent",
      "text": "Excellent! Is there anything else on the portal you'd like help navigating?"
    },
    {
      "role": "customer",
      "text": "No, I think I'm good now. Thank you so much for the warm welcome!"
    }
  ],
  "resolution": "Activation link had expired. New link sent and employee successfully activated account and logged into the HR portal."
}

How to Use
Load with Python
python

import json

with open("hr.json", "r", encoding="utf-8") as f:
    data = json.load(f)
    conversations = data["conversations"]

for conv in conversations:
    print(conv["problem"])

Load with Hugging Face Datasets
python

from datasets import load_dataset

dataset = load_dataset("ai-training-datasets/hr", split="train")
print(dataset[0])

Employee Personas Included

    frustrated user

    confused beginner

    impatient executive

    elderly user

    tech-savvy user

    calm and patient user

License

This dataset is released under the MIT License. You are free to use, modify, and distribute it for both research and commercial purposes.
Citation

If you use this dataset, please cite:
text

@dataset{ai_training_datasets_2026,
  title     = {AI Training Datasets: Human Resources Conversations},
  author    = {AI Training Datasets},
  year      = {2026},
  version   = {1.0},
  publisher = {Hugging Face}
}

## 💼 Commercial licensing

This dataset is available for free under the MIT License for non-commercial use.  
If you need a commercial license, custom datasets, or the full version with all 600 conversations, contact me at:

📧 cybernovasg@gmail.com